知识卡片
嵌入将对象映射为向量
用户和物品嵌入把各自对象映射为数值向量,用于比较、组合或作为预测网络输入。向量接近表示的是训练任务学到的关系,不自动等同现实语义。嵌入提供表示,后续评分或检索才决定如何形成推荐。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- · 查看原书依据
Embedding is a low-dimensional learning approach used to represent similar features in surrounding space [29]. Embeddings are generally used to find the nearest neighbor from surrounding space and visualize features in the lowest dimension. There are two kinds of embeddings: user and item, that are provided as inputs to machine learning models [30]. Moreover, CF is used to enable automatic learning of embeddings.
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第03章《深度推荐与评估指标》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C3.html)
- 结论依据:摘录中关于嵌入将对象映射为向量的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:Embedding is a low-dimensional learning approach used to represent similar features in surrounding space [29]. Embeddings are generally used to find the nearest neighbor from surrounding space and visualize features in the lowest dimension. There are two kinds of embeddings: user and item, that are provided as inputs to machine learning models [30]. Moreover, CF is used to enable automatic learning of embeddings.